R/isotonic.R

In mhweber/spsurvey: Spatial Survey Design and Analysis

################################################################################
# Function: isotonic
# Programmer: Tom Kincaid
# Date: August 15, 2001
#
#' Internal Function: Isotonic Regression
#'
#' This function performs isotonic regression of an input set of values so that
#' the output set of values is a nondecreasing sequence.  The output set of
#' values is truncated to the range: [minval, maxval].
#'
#' @param y Vector of the set of values on which to perform isotonic
#'   regression.
#'
#' @param minval Minimum value for the output set of values.
#'
#' @param maxval Maximum value for the output set of values.
#'
#' @return The revised set of input values
#'
#'   Other Functions Required:
#'      sorted - determine whether a set of values is a nondecreasing sequence
#' @section Other Functions Required:
#'   \describe{
#'     \item{\code{\link{sorted}}}{determines whether a set of values is a
#'       nondecreasing sequence}
#'   }
#'
#' @author Tom Kincaid \email{Kincaid.Tom@epa.gov}
#'
#' @export
################################################################################

isotonic <- function(y, minval, maxval) {

# Calculate additional required values

   n <- length(y)

# Truncate the output set of values to the range: [minval, maxval]

   y <- pmax(y, minval)
   y <- pmin(y, maxval)

# Perform isotonic regression

   while (!sorted(y)) {
      i <- 1
      while (i < n) {
         j <- i
         while (y[j] >= y[j+1] & j < n)
            j <- j+1
         if (i < j) y[i:j] <- mean(y[i:j])
         i <- j+1
      }
   }

# Return the result

   y
}